Silviculture conditions and wood properties of Samanea saman and Enterolobium cyclocarpum in 19-year-old mixed plantations

Transcription

1 Instituto Nacional de Investigación y Tecnología graria y limentaria (INI) Forest Systems (1), vailable online at ISSN: eissn: Silviculture conditions and wood properties of Samanea saman and Enterolobium cyclocarpum in 19-year-old mixed plantations M. F. Obando 1 and R. Moya 2 * 1 Escuela de Ingeniería Forestal. Instituto Tecnológico de Costa Rica. pdo Cartago, Costa Rica 2 Professor and Research. Escuela de Ingeniería Forestal. Instituto Tecnológico de Costa Rica. CIIBI. pdo Cartago, Costa Rica bstract im of study: The objective of this work is to examine wood properties of a mixed plantation of Samanea saman and Enterolobium cyclocarpum, combined with native species of Costa Rica. rea of study: Mixed plantation in dry tropical zone of Costa Rica Material and methods: It was measured by total diameter, heartwood, bark thickness, several physical wood properties (green density, specific gravity, volumetric, tangential and radial shrinkage and moisture content) and wood color. Main results: The largest of this diameter are reached when is planted with, and Swietenia macrophylla and Simarouba glauca. Meanwhile the largest total diameter and heartwood in E. cyclocarpum was found when this species is planted with Dalbergia retusa. Heartwood percentage, bark percentage, and some physical properties are not affected when or are planted with other species. However, parameters of wood color were affected mixture plantation of or with other species. The lowest difference between wood from mixed plantations and wood from trees growing in native forests is produced when is mixed with, but this last species is mixed with Hymenaea courbaril and produced wood with the lowest difference in wood color compared to trees growing in the natural forest. Research highlights: and tress planted in a mixed plantation produce variation in breast diameter, physical properties, heartwood percentage and wood color. Key words: plantation species; silvicultural condition; tropical species; wood properties. Introduction Enterolobium cyclocarpum (Guanacaste) and Samanea saman (Cenízaro) are important species used in furniture and craftwork industries. They have moderate wood density, attractive wood color, excellent workability properties and natural durability (Flores and Obando, 2003). Both species grow in dry tropical conditions, with temperatures from 18 to 25 C and elevation levels over 300 meters. (OFI/CTIE, 2003). However, these species have become scarce (Griscom and shton, 2010), resulting in consequent problems for the industry. Important efforts are being made towards reforestation. These Enterolobium cyclocarpum and Samanea saman species were tested in fast-growth condition plantations (Foroughbakhch et al., 2006). The excellent productivity and growth achieved with these species * Corresponding author: rmoya@itcr.ac.cr Received: ccepted: are considered important for fast-growing plantations in Costa Rica, Panama, Mexico and some frican and sian countries (Piotto et al., 2004). Mixed plantations (MP) have shown promising productivity (lice et al., 2004). Higher stand-level productivity in mixed species has been found with two kinds of species interactions: (1) complementary resource use between species that arise from development of a stratified canopy (and possibly root stratification); and, (2) Facilitative improvement in nutrition of a valuable timber species growing in combination with a nitrogen-fixing species (Kelty, 2006). Furthermore, mixed plantations combining fast and slowgrowth species show maximum productivity and profitability (Piotto et al., 2004). Just a few studies on wood properties of native forest species from mixed plantations are reported in literature and are limited to a reduced number of species and wood characteristics (Moya et al., 2009; Moya and Muñoz, 2010). However, the properties of wood from

2 Wood properties of Samanea saman and Enterolobium cyclocarpum 59 natural forest of species used in mixed plantations, such as and are well known (Carpio, 2003). wood has moderate specific gravity (from ), green density moderate values (0.77 g/cm 3 ), 0.45 g/cm 3 in air dry density, medium to fine texture, moderate shrinkage (2.6 % in tangential direction, 0.9% in radial direction, and 3.5% in volume), and excellent durability (révalo and Londoño, 2006; OFI/CTIE, 2003). Its heartwood is dark brown and its sapwood is white or pale yellow (OFI/ CTIE, 2003). On the other hand, has moderate mechanical resistance, moderate specific gravity (from 0.42 to 0.64), interlocked grain, moderate shrinkage and wood color difference between sapwood and heartwood (Carpio, 2003). The aim of the present study was to evaluate some physical properties of wood (green density, specific gravity, tangential, radial, and volume shrinkage, initial moisture content), aesthetic features such as color determined through Lab systems and heartwood presence from Enterolobium cyclocarpum and Samanea saman trees planted with other native Costa Rican species in mixed plantations. Methodology Geographic position This study was performed in mixed plantations from the Estación Experimental Forestal Horizontes in the Guanacaste Conservation rea (CG acronyms in Spanish) in Liberia, Guanacaste, Costa Rica (N W ) on a 30 hectare field. Mixed plantations are located in the Northern region of Costa Rica, with a dry tropical climate (Holdridge, 1987), where typical conditions are an average temperature of 24.0 to 24.5 C, annual precipitation of 1,500 to 2,000 mm, and a dry season from November to May. The soil has a loam to clay loam texture, it is moderately acid (ph 5-6) and the slope is less than 2%. Some sectors in the plantation are susceptible to flooding during the rainy season and mixed plantations T6, T7, T10 and T13 (Fig. 1) were affected by this fact. Characteristics of a mixed plantation Fourteen different mixed plantations located one next to the other were selected for the study (Fig. 1). Each mixed plantation was planted in 1 hectare. Main species were Enterolobium cyclocarpum (Jacq.) Griseb. and Samanea saman Merrill, Elmer Drew and they were planted in combination with other native Costa Rican species (Swietenia macrophylla King, Hymenaea courbaril L., Handroanthus impetiginosus (Mart. ex DC.) Mattos, lbizia adinocephala (Donn. Sm.) Record, Diphysa americana (Mill.) M. Sousa, Tabebuia rosea (Bertol.) Bertero ex.dc., Dalbergia retusa Hemsl. and Simarouba glauca DC.). The percentage of combination of different species on each mixed plantation (MP) and its respective codification are detailed in Fig. 1. T7 T8 +50% D. americana +50% T10 T11 +25% + 25% S. macrophylla 50% + 50% S. glauca T % T. rosea + 50% D. retusa +50% H. courbaril 50% +50% S. glauca T9 T12 T13 T2 + 25% H. courbaril +50% S. macrophylla 50% +50% H. courbaril T3 T6 + 50%. adinocephala +25% S. macrophylla T1 50% +50% S. macrophylla 50% +50% H. courbaril T4 T5 +50% T. impetiginosa Figure 1. Distribution of mixture plantation of or with different native species of 19 years-old in Costa Rica.

3 60 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), Plantation management The initial planting density was 1,111 trees/ha (3 3 m spacing) and each species were planted in alternating rows (inter cropping systems). t the moment of evaluation, the stands aged 19 years presented a density of between 495 and 575 trees/ha (Table 1). Trees were planted at the beginning of the rainy season (June and July). fter one year, it was necessary to replant some trees as trials. The trial had a sanitary thinning at age 6 and systematic thinning (33%) was applied at 11 years old. Each trial was pruned at the age of 6. Trees sampled Two plots (400 m 2 ) were established to characterize each MP (Fig. 1). The diameter at breast height and total height of all trees of or within each plot were measured. Three trees were selected to represent all the diameter range in each MP. The harvested trees had straight boles with normal branching and without any visible disease or pest symptoms. Before cutting the trees, their north-facing side was marked. Then, a 200 cm long log from the basal part of the trunk was obtained. Two 4 cm thick discs (cross section) were obtained in the upper part of this log (200 cm from the soil line). Heartwood and bark percentage determination First, on each disc, a cross-sectional line was drawn from north to south passing through the pith and another was marked perpendicularly to the first, from east to west. Total diameter, diameter without bark, Table 1. Morphologic characteristics of trees of and from mixture plantation with different native species of 19 years-old in Costa Rica Mixture plantation d (cm) d H (cm) b (cm) T BCD (±0.94) BC (±2.54) 1.01 (±0.17) T DC (±0.35) BC (±1.76) 1.16 (±0.19) T B (±1.82) (±1.85) 0.85 (±0.12) T (±2.21) B (±0.88) 1.06 (±0.34) T BC (±1.79) B (±3.09) 0.90 (±0.07) T D (±1.46) C (±2.15) 1.02 (±0.14) verage (±3.42) (±4.11) 1.02 (±0.18) T B (±5.37) B (±4.63) 0.77 (±0.13) T B (±0.08) B (±0.83) 0.85 (±0.13) T B (±1.59) B (±0.62) 0.82 (±0.22) T B (±2.22) (±1.65) 0.79 (±0.15) T B (±0.89) B (±1.00) 0.77 (±0.12) T (±2.07) (±1.72) 1.03 (±0.15) T B (±2.21) B (±1.61) 0.71 (±0.14) T B (±1.12) (±1.77) 0.80 (±0.16) T B (±1.25) B (±2.07) 0.74 (±0.09) verage (±4.27) (±3.56) 0.81 (±0.15) d T : total diameter. d h : heartwood diameter. b: bark thick. T1: 50% + 50% Swietenia macropylla. T2: 25% + 25% Hymenaea courbaril + 50% S. macrophylla. T3: + 50% Hymenaea courbaril. T4: 50% + 50% Hymenaea courbaril. T5: + 50% Tabebuia impetiginosa. T6: + 25% lbizia adinicephala + 25% S. macrophylla. T7: + 50% Diphysa americana. T8: + 50% Tabebuia rosea. T9: + 50% Dalbergia retusa. T10: + 50%. T11: + 25% + 25% S. macrophylla. T12: + 50% H. courbaril. T13 y T14: 50% + 50% S. glauca. Note: Different letters for each general characteristic indicate that values are statistically different at a 99% confidence level and same letters for each general characteristic indicate that values are not statistically different.

4 Wood properties of Samanea saman and Enterolobium cyclocarpum 61 pith diameter, and heartwood diameter (see Fig. 1 in Moya and Muñoz, 2010) were measured on a crosssectional line drawn in two directions (north-south and east-west). Means for all diameters were calculated as the average of two cross-sectional measurements on each stem section. Next, the total heartwood and pith cross-sectional areas were calculated as a geometric circle, and the bark content was determined for the difference between the total area and the area without bark. Wood physical properties The following wood physical properties were determined: radial shrinkage (RS), tangential shrinkage (TS) and total volume shrinkage (VS), wood density at green condition (GD), initial or green moisture content (IMC) and specific gravity (SG). They were determined according to STM D-143 (STM, 2003a). second 4 cm thick diametric section was cut into the cross section from north to south and it was cut at pith, obtaining two diametric sections. North diametric sections were re-cut to obtain 2.0 cm thick samples tangential to the growth rings and in bark-pith direction. The volume and weight of each sample were measured. The volume was defined as the volume of water it displaced when submerged, according to STM D standards (STM, 2003b). fterwards, the GD, SG and VS were determined. GD was calculated as weight divided by volume, while IMC was calculated as the difference between green and dry weight and divided by dry weight; and both values are expressed as percentages, according to STM D (STM, 2003a). SG was calculated as the oven-dry weight divided by volume in green condition. VS was determined as the difference between green and oven-dry volume and divided by green volume. The east and west sides of the disk were used to measure RS and TS. Refer to Moya and Muñoz (2010) for details regarding type of sample cut. They were determined as the difference between green and oven-dry dimensions, and divided by green dimension. Wood color cross section used for heartwood and bark determination was also used for wood color determination. 3 cm wide diametric section was extracted for all sample trees. These samples were conditioned at 22 C and at a relative humidity of 66% in order to obtain a 12% in equilibrium moisture content. Wood color was measured along radial transect at four distances from the pith (near the pith, midway between pith and sapwood/heartwood boundary, sapwood/heartwood boundary, and halfway into sapwood thickness). Wood color was measured on the tangential side of the block according to ISO 7724/ and ISO 7724/ (ISO 1984). HunterLab MiniScan XE Plus spectrophotometer was used. Measurements were taken at room temperature and color characteristics were determined using the CIELab system, which determines lightness, redness and yellowness (parameters L*, a*, and b* respectively). Measurement range was 400 to 700 nm, with a 13 mm aperture at the measurement point. The specular component (SCI mode) was included to observe reflection at a 10º angle, which is normal for the specimen surface (D65/10), as well as a 2 field of vision (standard observer, CIE 1931), and standard D65 illumination (corresponds to daylight at 6,500 K). Color difference ccording to STM D2244 (STM, 2003c), the color difference between two wood samples measured using the CIELab system can be expressed as an index called ΔE* (Equation 3). In this study, heartwood difference was calculated (ΔE*) between different mixed plantations of the same species ( or ). lso, the ΔE* was calculated between wood color of trees from different mixed plantations and wood color of trees from natural forests. The calculation of ΔE* heartwood was deemed important in order to assess the color quality of different mixed plantations. Therefore, the standard heartwood color was obtained from samples from natural forest in Xylariorum Víctor Rojas at the Costa Rica Institute of Technology. The wood samples were TECw-29 for and TECw-83 for. Values L*, a*, and b* were averaged out for this pattern and then ΔE* was calculated in accordance with STM D2244 (STM, 2003c) using equation 1: [1] where: ΔL*=L* ΔL* P, Δa* = Δa* M Δa* P, Δb* = b* M Δb* P, M = value of sample color coordinate of a mixture plantation and p =value of pattern color coordinate or other mixture plantation.

5 62 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), Statistical analysis descriptive analysis was first conducted to evaluate the normal distribution of data, homogenous variance, and absence of extreme data in the heartwood and bark percentage, physical properties and color parameters. dditionally, average values, standard deviation, and variation ranges were calculated using the SS UNIVRITE procedure (version 8.1 for Windows). fter that, analysis of variance (NOV) was used to determine the causes for wood properties measured in this study, then the Tukey test was applied to establish differences between the averages of wood properties from different mixture plantations. Regression analysis was used to determine the relationships of color coordinates (L*a*b*) or specific gravity and pith distance. SS (SS Institute Inc.) and STTISTIC 6.0 (Statsoft Inc.) programs were used for the statistical computations. Results Morphologic characteristics at 200 cm in height The average diameter from total diameter and heartwood (Table 1) of are highest when this species was planted next to (50%-50%, known as T10); with and S. macrophylla (50%-25%-25%, named as T11) and when it is planted together with S. glauca (50%-50%, known as T13). The total diameter varied from 35.0 to 38.0 cm and heartwood diameter ranged from 24.0 to 28.5 cm (Table 1). For other MP of, the total diameter varied from 29.0 to 31.5 cm and from 17.0 to 20.5 cm in heartwood diameter (Table 1). For, the highest diameter was found in MP of this species planted with D. retusa (50-50%, named T9), but heartwood diameter was not statistically different among MPs (Table 1). The highest value of heartwood percentage (HWP) in appeared when this species was mixed with (T10). However, statistical differences with other MP with and other native species (Fig. 2a) were not found with the exception of MP of and S. glauca (T14), which presented the lowest values of HWP (Fig. 2a). On the other hand, MP with produced lower values of HWP than. Meanwhile, MP of and other native species of Costa Rica produced no significant differences in HWP (Fig. 2a). Regarding BP or pith diameter, it was found that MP with or and other native species did not produce statistical differences in BP and pith diameter (Table 1 and Fig. 2b). Physical properties Table 2 shows the average values for physical properties (GD, IMC and different types of shrinkages) at 200 cm for MP of or. IMC varied from 99 to 124% in MP of, meanwhile IMC in MP of was slightly higher, the average varied from 111 to 150% (Table 2). It was found that GD ranged from 0.95 to 1.03 g/cm 3 in S. saman. The highest values were found in E. cyclocar- a) b) B B B B B T1 T3 T4 T10 T11 T13 T14 T2 T5 T6 T7 T8 T9 T10 T11 T T1 T3 T4 T10 T11 T13 T14 T2 T5 T6 T7 T8 T9 T10 T11 T12 Heartwood (%) Figure 2. Heartwood () and bark (B) proportions in trees of and from mixture plantation with different native species of 19 years-old in Costa Rica. For treatment name see legend ot table 1 and Fig. 1. Different letters for each general characteristic indicate that values are statistically different at a 99% confidence level. Bark (%)

6 Wood properties of Samanea saman and Enterolobium cyclocarpum 63 Table 2. Physical properties of wood of trees of and from mixture plantation with different native species of 19 years-old in Costa Rica Mixture plantation Initital moisture content (%) Shrinkage Green density (g/cm 3 ) Volume Radial Tangential (%) (%) (%) T1 99 (±37) 0.95 (±0.14) 6.27 (±1.20) 2.76 (±0.53) 3.72 (±0.62) T3 101 (±26) 1.03 (±0.11) 6.59 (±1.10) 3.47 (±0.70) 4.64 (±0.51) T4 100 (±36) 0.99 (±0.14) 6.93 (±1.41) 2.85 (±0.54) 4.94 (±1.70) T (±48) 0.98 (±0.13) 6.33 (±1.32) 2.51 (±0.46) 3.80 (±0.86) T (±38) 0.96 (±0.13) 6.75 (±1.34) 2.92 (±0.53) 3.06 (±1.16) T (±37) 1.01 (±0.11) 6.44 (±1.09) 2.81 (±0.47) 4.00 (±0.15) T (±29) 1.00 (±0.12) 6.52 (±1.15) 2.23 (±0.71) 4.40 (±0.88) verage 105 (±38) 0.99 (±0.13) 6.55 (±1.25) 2.79 (±0.58) 4.08 (±0.98) T2 149 (±69) 0.80 (±0.28) 2.91 (±1.20) 1.10 (±0.74) 2.84 (±0.41) T5 138 (±72) 0.76 (±0.26) 3.25 (±1.09) 2.20 (±0.51) 3.18 (±0.10) T6 125 (±55) 0.82 (±0.23) 3.63 (±1.16) 1.17 (±0.90) 3.28 (±0.80) T7 130 (±64) 0.86 (±0.23) 3.80 (±1.17) 2.07 (±0.82) 3.71 (±0.26) T8 122 (±66) 0.76 (±0.27) 2.88 (±0.85) 1.20 (±0.10) 3.61 (±0.36) T9 107 (±57) 0.70 (±0.26) 3.63 (±1.17) 1.95 (±1.19) 4.52 (±0.11) T (±71) 0.72 (±0.24) 5.26 (±1.23) 1.97 (±0.72) 3.91 (±1.95) T (±70) 0.84 (±0.27) 3.32 (±0.92) 1.97 (±0.63) 3.46 (±0.21) T (±55) 0.83 (±0.25) 2.91 (±0.92) 2.14 (±0.55) 3.55 (±0.50) verage 125 (±65) 0.78 (±0.26) 3.66 (±1.13) 1.74 (±0.75) 3.56 (±0.76) For treatment name sees legend of Table 1. Different letters for each general characteristic indicate that values are statistically different at a 99% confidence level and same letters for each general characteristic indicate that values are not statistically different. Specific gravity T1 T3 T4 T10 T11 T13 T14 T2 T5 T6 T7 T8 T9 T10 T11 T12 Figure 3. Variation of specific gravity of and from mixture plantation with different native species of 19 years-old in Costa Rica. For treatment name sees legend of Table 1 and Fig 1. pum, varying from 0.70 to 0.86 g/cm 3. For SG, it was determined that MP of presented values from 0.48 to 0.51 and these values are higher than values measured in, varying from 0.31 to 0.38 (Fig. 3). Finally, different shrinkages show same performance of SG, the highest shrinkage (VS, TS and RS) were presented in (Table 2). n important result from the NOV analysis was that no physical properties were statistically different among different MP of or (Table 2 and Fig. 3). Radial variation of physical properties shows that both species evaluated in different MPs presented different performance from pith to bark direction. SG increased from pith to bark direction in trees from MP, the lowest values are presented near the pith and afterwards the SG increases with distance from pith (Fig. 4a). However, SG did not present any relation to distance from pith, SG was slightly constant (Fig. 4b). IMC decreased in both species, however the variation of IMC of was irregular and high variation (Fig. 4c), meanwhile the decreasing in was uniform (Fig. 4d). Green density in both species was constant from basal tree part until 50% of distance from pith to bark, afterwards this physical properties decreased with distance from the pith (Fig. 4e and 4f). Finally, volumetric shrinkage

7 64 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), a) 0.65 b) y = x x R 2 = c) 300 d) Specify gravity Initial moisture content (%) y = e 0.008x R 2 = y = e 0.057x R 2 = e) Green density (g/cm 3 ) y = e 0.007x R 2 = y = x R 2 = f) g) i) Total volume shrinkage (%) y = x R 2 = y = x x R 2 = Distance relative of pith (%) Distance relative of pith (%) Figure 4. Variation of specific gravity, initial moisture content, green density and total volumen shirinkage in trees of (a) and (b) from mixture plantation with different native species of 19 years-old in Costa Rica.

8 Wood properties of Samanea saman and Enterolobium cyclocarpum 65 was not relation with pith distance in (Fig. 4g), but there was an increasing with distance from pith in (Fig. 4h). Wood color The color composition of S. Saman or wood can be described using the combination of different tonalities of lightness (L*), redness (+a*) and yellowness (+L*). Sapwood color was different than heartwood in trees from MP of S. Saman or. Wood color of sapwood was whitish or yellowish and heartwood was reddish or brownish. This difference was produced by a statistical difference in two of three color parameters. Heartwood of both species presented the lowest values of L* (lightness) and they were statistically different from sapwood. The a* color parameter (redness) was higher in heartwood than sapwood. statistical difference in b* color parameter between heartwood and sapwood from both planted species was not found (Fig. 5). NOV shows that three color parameters (L*, a* y b*) of sapwood of two species from different MP was Values B B L* a* b* L* a* b* Heartwood Sapwood Figure 5. Wood color parameters of sapwood and heartwood of trees of and from mixture plantation with different native species of 19 years-old in Costa Rica. For treatment name sees legend of Table 1. Different letters for each general characteristic indicate that values are statistically different at a 99% confidence level. not found statistically different among them (Table 3). In relation to heartwood, it was demonstrated that wood from trees growing in different MP, lightness parameter (L*) and redness parameters (a*) were not different in different MP. But yellowness tonality (b*) was affected when was planted with other B B Table 3. Wood color parameters measured by CIE Lab of and from mixture plantation with different native species of 19 years-old in Costa Rica Mixture plantation Heartwood Sapwood L* a* b* L* a* b* T (±4.78) 7.65 (±1.65) B (±0.89) (±4.43) 2.55 (±1.54) (±1.38) T (±2.52) 6.98 (±0.88) B (±1.73) (±5.94) 2.47 (±1.51) (±0.69) T (±3.73) 7.90 (±0.93) B (±3.06) (±0.67) 2.44 (±0.12) (±1.46) T (±6.38) 7.22 (±2.20) B (2.99) (±3.26) 2.12 (±1.18) (±4.33) T (±1.59) 7.73 (±0.24) B (0.97) (±0.79) 3.16 (±0.27) (1.52) T (±3.24) 7.04 (±0.63) 20.5 B (±2.48) (±3.15) 2.71 (±0.95) (±5.48) T (±3.11) 6.79 (±0.93) B (±1.02) (±3.98) 2.37 (±0.84) (±2.49) E. Cyclocarpum T B (±2.60) 6.93 B (±0.32) (±0.44) (±1.12) 0.98 (±0.47) (±0.68) T B (±0.85) 8.27 B (±0.57) (±1.65) (±5.15) 1.45 (±0.26) (±3.43) T B (±7.73) 7.05 B (±0.98) (±1.45) (±2.45) 1.18 (±0.26) (±1.41) T B (±5.47) 7.72 B (±0.53) (±2.07) (±7.91) 1.04 (±0.59) (±1.17) T B (±1.23) 8.86 (±0.81) (±1.56) (±1.06) 1.55 (±0.54) (±2.12) T B (±4.56) 8.27 B (±0.97) (±3.29) (±2.80) 1.34 (±0.94) (±2.04) T (±4.97) 7.87 B (±1.81) (±0.90) (±0.87) 1.10 (±0.44) (±3.05) T B (±2.14) 8.29 B (±0.90) (±0.53) (±1.30) 1.19 (±0.21) (±1.26) T B (±2.51) 7.62 B (±1.51) (±0.33) (±3.68) 1.33 (±0.20) (±2.02) L*: lightness from black (0) to white (100). a*: redness (+) to greeness ( ). b*: yellowness (+) to bluness ( ). For treatment name sees legend of Table 1. Different letters for each general characteristic indicate that values are statistically different at a 99% confidence level and same letters for each general characteristic indicate that values are not statistically different.

9 66 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), tropical species (Table 3). Wood from trees planted with S. glauca (T13 and T14) had higher b* than wood from trees planted with and S. macrophylla (T11). In relation to E. cyclocarpum, it was found that b* color parameters were not affected by different MP. But for L* and a* parameters some differences were established. Wood from trees growing in MP of and S. saman (T10) presented higher values than wood from trees growing in MP of and D. americana (T7). Furthermore, wood from these MP were not different from other MP. The a* color parameter of heartwood was highest in wood from trees growing in MP planted with and T. rosea (T9) and this value was significantly different than wood from trees in MP of, H. courbaril and S. macrophylla (T2) (Table 3). Wood color parameters of S. Saman and varied according to distance from the pith (Fig. 6). L* and b* color parameters (lightness and yellowness respectively) decreased from near the pith up to the heartwood boundary, afterwards the value increased in the sapwood section (Fig. 6a and 6c), with the exception of b* parameter of, which decreased in the sapwood section (Fig. 6c). On the other hand, the redness parameter (a*) tends to remain with distance from pith in the heartwood section, and then this value decreased in the sapwood part (Fig. 4b). This variation of wood color parameters in the wood in inner part of heartwood produced a lighter color than the wood near the heartwood boundary, which means that the wood becomes increasingly dark. Discussion Trees from MP with higher HWP are considered an advantage because their wood is an important factor for determining specific uses such as furniture and decorative veneers, both very important marketing attributes (Mazet and Janin, 1990) associated with high decay resistance as well (Moya and Berrocal, 2010). ccording to that, MP of with S. glauca (T14) is less favourable because it produced trees with the lowest HWP values. However, this result must be considered with care, because other MP with and S. glauca (T13) located near T14 (Fig. 1) did not present statistical differences in HWP and total diameter regarding other MP of (Table 1). The difference found in total diameter and heartwood in two MPs a) Parameter L* b) Parameter a* c) Parameter b* Pith Middle Heartwood Sapwood heartwood boundary Figure 6. Relationship between distance from the pith and L*, a*, b* for wood of trees of and from mixture plantation with different native species of 19 years-old in Costa Rica. of with S. glauca (T13 and T14) may be due to the difference in site. The T14 MP was planted on a site susceptible to flooding. Hocking and Islam (2011) studied the growth of in natural forests and they found that the diameter was reduced in sites susceptible to flooding. HWP in all MPs was lower than 50%. However this percentage is common in tropical species growing in fast-growth plantations. For example, Moya and Muñoz (2010) and Moya et al. (2009) found similar or higher values of HWP than our results in Terminalia amazonia

10 Wood properties of Samanea saman and Enterolobium cyclocarpum 67 (J. F. Gmel.) Exell, T. oblonga (Ruiz & Pav.) Steud., Vochysia guatemalensis Donn. Sm., and Bombacopsis quinata (Jacq.) Dugand, except cacia mangium Willd. and Cupressus lusitanica Mill. in pure plantations, which resulted in 65% and 70% respectively. nother important result of this study was that HWP was not affected in any MP of or (Fig. 2b), but total diameter and heartwood diameter was higher in some MPs of these species (Table 1). For example, and (T10) MP mixing with and S. macrophylla (T11) or with S. galuca (T13) produced the highest total diameter (Table 1). Lumber recovery Factor (LRF) is associated to log diameter (Steele 1984). In general, the overall conversion factor, volume yield, and output of saw milling increased with increasing log diameter. In other words, lumber recovery is higher for logs with larger diameters (Pnevmaticos et al., 1970; Wang et al., 2003). Therefore, trees from MP with the highest total diameter will produce higher LRF than trees from MP with the lowest total diameter. On the other hand, BP found in two native species of Costa Rica in MP conditions corresponds to other native species planted in fast-growth conditions in this country. Moya and Muñoz (2010) reported a range in variation of 8.7 to 20.1% for lumber from trees coming from fast-growth plantations. Bark of trees has different physiological properties; they are related to the ecology of the different tree species and provide different habitats for bark living arthropods (Nicolai 1986) or genetic control (Zhong et al., 2009). The thickness of the bark correlates with factors such as moisture availability, tree age, genetic and diameter classes (Sonmez et al., 2007). The lack of difference in bark thickness (Table 2) or percentage of bark (Fig. 1b) among mixed plantations can be attributed to a genetic control of bark in each species (Zhong et al., 2009). SG average was This result agrees with other reports. OFI/CTIE (2003) reported values ranging from 0.42 to However, different shrinkage values (VS, TS and RS) disagreed with our results. Meanwhile, revalo and Londoño (2006) reported a range from 0.35 to 0.60 for trees growing in natural forests in Colombia. The SG found for this species in our study (from 0.35 to 0.38) is included in révalo and Londoño s range. Tamarit and Fuentes (2003) studies for also from natural forest, found values of SG (0.35) and volume shrinkage (8.40%) similar to our results; however, they reported values in IMC (220%) higher than values found in different MP. This divergence between our results and these references suggests a probable influence from the site, environmental conditions or juvenile presence on wood quality. Zobel and Van Buijtenen (1989) suggest that large structure variations are produced by changes in climate, site, juvenile wood and management characteristics. The differences between those references and our findings may be due to: (1) trees were harvested in fast-growth condition, (2) trees age is only 19 years old, and (3) high juvenile wood proportions. SG can increase with tree age or pith distance in trees from all MPs, but mixed plantations did not showed any relation with pith distance. SG is a measurement of the amount of wood fiber cell walls and void space, and therefore this parameter determines the amount of water in wood (Moya and Tomazello, 2007). In hardwood species, the wood density variation depends on the amount or proportion of different cell types and spatial arrangements. Four different cell types are present in hardwood, each more or less specialized in structure and restricted in function; rays; vessels; fibers; axial parenchyma. In addition, many molecular and physiological changes occur in the vascular cambium during the aging process (Plomion et al. 2001). Thus, this explains the SG increase with the aging of and we can associate aging with thick-walled fibers and vessel diameter, but a vessel frequency decrease in. Whereas anatomical variations in are different and they may be affecting the typical development of SG with tree age, anatomical variations of fiber, vessels or ray tissue affect SG differently (Moya and Tomazello, 2007). Wood with low uniformity of SG from pith to bark is one of the greatest problems for the industrial face. Echols (1973) attributed wood density uniformity degree as a limiting factor for the end use of the wood considering the effect in the sawing, drying, machining, gluing, finishing and other wood quality. ccording to this, the log of, with low uniformity in SG across transversal sections, may present better performance in sawing, drying, machining, gluing and finishing processes than the log. The difference in color between heartwood and sapwood of and is due to the process of heartwood formation, with the extractives produced accumulating in the heartwood, resulting in color (Gierlinger et al., 2004). The process depends on

11 68 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), several factors, such as tree age, soil composition and location within a tree. Some precursors of the synthesis of extractives are located in sapwood tissue, but the content is lower than in heartwood (Gierlinger et al., 2004). For various other species, tree age or pith distance has shown significant correlations with color parameters, similar to the two native species of Costa Rica. For example, Gierlinger et al. (2004), studying hybrids of Japanese larch, have shown that old trees have significant redder hue (a*) than young trees. Klumpers et al. (1993) found that the heartwood became increasingly redder (a*) with age in Quercus robur. Moya and Berrocal (2010) studying wood color variation for Tectona grandis, showed that * L* and a* were negatively and positively respectively correlated with pith distance in heartwood, but not for b*. In this case, there are differences between two native species, a* color parameters were maintained and b* parameters decreased across heartwood in and, but for T. grandis this relationship was different, a* increased and b* was maintained across heartwood sections (Moya and Berrocal, 2010). The wood color difference index ΔE* (Equation 1) is expressed as a distance between two points in the color coordinate system, a quadratic addition of the difference in each coordinate (Gonnet 1993). Cui et al. (2004) mentioned the value of color change (ΔE*) to define the levels of perceived difference in color. ΔE* varying for 0 to 0.5 is defined as negligible, between 0.5 and 1.5 is slightly perceivable, from 1.5 to 3.0 is noticeable, from 3.0 to 6.0 is appreciable, and from 6.0 to 12.0 is very appreciable. Consequently for this study, a* value higher than 1.5 units in wood color difference index (ΔE*) from two different mixed plantations will produce a difference noticeable or appreciable in visual perception. Table 4 includes a matrix comparing the color difference index ΔE* among mixed plantations of and. From this table we can observe that mixed plantations of these two tropical species produce different wood color in heartwood. It was found that the ΔE* values in mixed plantations were higher than 1.5 units in most comparisons of two MPs except in comparison between T4 with T1 ( with S. macrophylla plantation and with H. courbaril plantations, respectively) and T3 with T13 ( with H. courbaril and S. saman with S. glauca plantation respectively). The other color comparisons between mixed plantations with showed that wood color difference between different mixed plantations is cataloged as noticeable and appreciable (ΔE* ranged from 1.56 to 4.88). nother important result is that wood from mixed plantations produced large wood color differences in relation to wood from tree growing in natural forests. ΔE* was higher than 35 in all comparisons (Table 4). The mixed plantations of and produced trees with the lowest difference in wood color compared to wood color from trees growing in the natural forest (ΔE* = 35.71). heartwood color comparison between different MP (Table 4) showed that MP T8 (E. cyclorcarpum with T. rosea) and T5 ( with T. impetiginosa) did not produced any color difference than other MPs. No difference was found between T8 and T5 (E. cyclorcarpum with T. rosea and with T. impetiginosa plantation, respectively), T9 and T5 (E. cyclorcarpum with D. retusa and with T.impetiginosa plantation, respectively), T9 and T8 (E. cyclorcarpum with D. retusa and with T. rosea), T11 and T5 (E. cyclorcarpum with and S. macrophylla and with T.impetiginosa plantation, respectively) and T11 and T9 (E. cyclorcarpum with and S. macrophylla and with D. retusa plantation, respectively). Whereas, wood from T2, T6, T7, T10 and T12 MPs was different among them, and different than T5, T8, T9 and T11. It is important to observe that trees from T10 MP (E. cyclorcarpum with ) produced wood with large differences in heartwood color compared to wood from other mixed plantation, especially T5, T7, T8 and T9. Trees from T12 produce too large wood color differences than T2 and T10 (Table 4). Other color comparisons showed that color differences between MPs are catalogued as noticeable and appreciable (ΔE* ranged from 1.56 to 4.88). Furthermore, comparisons between S. samam (?) and wood from mixture plantation produced large wood color difference in relation to wood from trees growing in natural forests. ΔE* varied from to (Table 4). ccording to these results, mixture plantation T7 (E. cyclocarpum with D. americana) produced wood with the lowest difference in wood color extracted from trees growing in the natural forest (ΔE* = 25.49). Wood color differs widely among species as well as within a single tree (Liu et al. 2005). Pith distance is the main factor of heartwood color variation within a tree (Klumpers et al., 1993; Moya and Berrocal, 2010). Some tropical species growing in fast-growth pure

12 Wood properties of Samanea saman and Enterolobium cyclocarpum 69 plantation agreed with the variation of color parameters in relation to pith distance for L* and b* parameters (Moya and Berrocal, 2010). nd these variations in color parameters are attributed to extractive content, especially to phenol extractive types (Gierlinger et al., 2004). Silvicultural implications Higher diameter is reached when and E. cyclocarpum are planted together in a MP, specifically when they are planted in a relation of 50%-50%. nother important mixture is when is planted with and S. macrophylla (50%-25%- 25%) and with S. glauca (50%-50%). ccording with these results, these MPs are the best combination for production of and. However, the combination of with D. retusa (50%-50%) reached the largest heartwood diameter in was found when this species is planted with D. retusa (50%-50%). Then, if these species are important for its heartwood properties, the management of MP may be oriented for heartwood increment. Silvicultural practices such as the usage of spacing and thinning help to heartwood proportions (Gartner, 2005). nother important aspect is that the physical properties are lightly affected when or are planted with other native Costa Rican species. Therefore the management can be oriented in the heartwood increment and wood properties were not affected. On the other hand, if we want to produce wood with little wood color difference with wood from trees growing in native forests, the management of MP of or with other tropical species must be different. For example, large heartwood proportion is reached when is planted next to (50-50%), but this combination produce wood with large color difference in relation to wood from native species. Therefore, this management help productivity but not wood color. Conclusion and planted in MP with other native tropical species in Costa Rica produce some variations in total and heartwood diameter of a tree. The largest of these diameters is reached when S. saman is planted with (50%-50%), with and S. macrophylla (50%-25%- 25%) and with S. glauca (50%-50%). Meanwhile, the largest total and heartwood diameter in was found when this species is planted with D. retusa (50%-50%). Large variations in these tree parameters were also found. Heartwood percentage, bark percentage, some physical properties (IMC, SG, GD and different shrinkage) are not affected when or are planted with other native Costa Rican species. However, wood color, an important wood characteristic for commercial use, was affected when or were planted with other tropical species. The lowest difference between wood from mixed plantations and wood from trees growing in native forests is produced when is planted next to (50-50%). But this last species, when mixed with H. courbaril, produced wood with the lowest difference in color than wood extracted from trees growing in a natural forest. cknowledgement The authors wish to thank Vicerrectoría de Investigación y Extensión of Instituto Tecnológico de Costa Rica (ITCR) for finnantial sopport, and Milena Gutiérrez Leitón of Estación Experimental Forestal Horizonte for raw materials and facilities for the study. References lice F, Montagnini F, Montero M, Productividad en plantaciones puras y mixtas de especies forestales nativas en la Estación Biológica La Selva, Sarapiquí, Costa Rica. gronomía Costarricense 28(2): révalo R, Londoño, Manual para la identificación de maderas que se comercializan en el departamento del Tolima. Universidad de Tolima Colombia, Colombia. 105 pp. STM, 2003a. Standard D Test Methods for Small Clear Specimens of Timber, STM International, West Conshohocken, P-US. 12 pp. STM, 2003b. Standard D Standard Test Methods for Specific Gravity of Wood and Wood-Based Materials, STM International, West Conshohocken, P-US. 11 pp. STM, 2003c. Standard D Standard Practice for Calculation of Color Tolerances and Color Differences from Instrumentally Measured Color Coordinates, STM International, West Conshohocken, P-US. 8 pp.

13 70 M. F. Obando and R. Moya / Forest Systems (2013) 22(1), Carpio M, Maderas de Costa Rica 150 especies forestales. Editorial de la Universidad de Costa Rica, San José, Costa Rica. 151 pp. Cui W, Kamdem P, Rypstra T, Diffuse reflectance infrared fourier transform spectroscopy (DRIFT) and color changes of artificial weathered wood. Wood Fiber Sci 36: Echols R, Uniformity of wood density assessed from x-ray of increment cores. Wood Sci Tech 7: Flores E, Obando G, Árboles del trópico húmedo: importancia socioeconómica. Editorial Tecnológica de Costa Rica, Cartago, Costa Rica. 315 pp. Foroughbakhch R, lvarado M, Hernández JL, Rocha, Guzmán M, Establishment, growth and biomass production of 10 tree woody species introduced for reforestation and ecological restoration in northeastern Mexico. Forest Ecol Manag 235: Gierlinger N, Jacques D, Gradner M, Wimmer R, Schwanninger M, Rozenberg P, Pâques LE, Colour of larch heartwood and relationships to extractives and brown-rot decay resistance. Trees 18: Gonnet J, CIELab measurement, a precise communication in ower colour: an example with carnation (Dianthus caryophyllus) cultivars. J Horticultural Sci Biotech, 68: Griscom HP, shton MS, Restoration of dry tropical forests in Central merica: a review of pattern and process. Forest Ecol Manag 261: Hocking D, Islam K, Trees in Bangladesh paddy fields. 2. Survival of trees planted in crop fields. groforestry Syst 31: Holdridge L, Life zone ecology. Tropical Science Center San José, Costa Rica. 415 pp. ISO Standard 7724:2, Paints and Varnishes Colorimetry Part 2: Colour,easurement, 1 st ed. Ref. No. ISO 7724:2-1984(E). 28 pp. Klumpers J, Janin G, Berker M, Lévy G, The Influences of age, extractive content and soil water on wood color in oak: the possible genetic determination of wood color. nnals For Sci 50: Kelty MJ, The role of species mixtures in plantation forestry. Forest Ecol Manag 233(2-3): Liu S, Loup C, Gril J, Dumonceaud O, Thibaut, Thibaut B, Studies on European beech (Fagus sylvatica L.). Part 1: Variations of wood color parameters. nnals For Sci, 62: Mazet JF, Janin G, La qualité de l aspect des placages de Chênes (Quercus petraea and Quercus robur): mesures de couleur et critères de d appréciation des professionnels français et italiens. nnals For Sci 47: Moya R, Berrocal, Wood colour variation in sapwood and heartwood of young trees of Tectona grandis and its relationship with plantation characteristics, site, and decay resistance. nnals For Sci 67(1): Moya R, Tomazello M, Relationship between anatomical features and intra-ring wood density profiles in Gmelina arborea applying X-ray densitometry. Cerne, 13: Moya R, Leandro L, Murillo O, Wood characteristics of Terminalia amazonia, Vochysia guatemalensis and Hyeronima alchorneoides planted in Costa Rica. Revista Bosques 30(2): Moya R, Muñoz F, Physical and mechanical properties of eight species from fast-growth plantation in Costa Rica. J Trop Forest Sci 22: Nicolai V, The bark of trees: thermal properties, microclimate and fauna. Oecologia 69: OFI/CTIE, Árboles de Centroamérica: un manual para extensionistas. Oxford Forestry Institute/Centro gronómico Tropical de Investigación y Enseñanza. Turrialba. CR. 569 pp. Piotto D, Víquez E, Montagnini F, Kanninen M, Pure and mixed forest plantations with native species of the dry tropics of Costa Rica: a comparison of growth and productivity. Forest Ecol Manag 190(2-3): Plomion C, Leprovost G, Stokes, Wood Formation in Trees. Plant Physiology 127: Pnevmaticos SM, Petro FJ, Flann IB, Yellow birch log attributes and sawmill productivity. Forests Prod J 20(8): Steele P, Factors determining lumber recovery in sawmilling. Gen. Tech. Rep. FPL-39. Madison, WI: US Department of griculture, Forest Service, Forest Products Laboratory. 25 pp. Sonmez T, Keles S, Tilki F, Effect of aspect, tree age and tree diameter on bark thickness of Picea orientalis. Scandinavian J Forest 22: Tamarit JC, Fuentes M, Parámetros de humedad en 63 maderas latifoliadas mexicanas en función de su densidad básica. Revista Chapingo Serie Ciencias Forestales y del mbiente 9(2): Wang S, Lin C, Chiu M, Effects of thinning and pruning on knots and lumber recovery of Taiwania (Taiwania cryptomerioides) planted in the Lu-Kuei area. J Wood Sci 49: Zhong LJ, rnold R, Xiang DY, Jian Y, Jian C, Genetic variation in growth, wood density and bark thickness of Eucalyptus dunnii. J Forest 22: Zobel B, Van Buijtenen B, Wood variation: its causes and control. Springer Verlag, New York, US. 346 pp.